Chapter Contents |
Previous |
Next |
PROBPLOT Statement |
See CAPPROB3 in the SAS/QC Sample Library |
This example is a continuation of "Creating Lognormal Probability Plots". Figure 9.4 shows that a lognormal distribution with shape parameter is a good fit for the distribution of DIAMETER in the data set MEASURES.
The lognormal distribution involves two other parameters: a threshold parameter and a scale parameter .See Table 9.13 for the equation of the lognormal density function. The following statements illustrate how you can request a diagonal distribution reference line whose slope and intercept are determined by estimates of and .
symbol v=dot height=3.5pct; legend2 frame cframe=ligr cborder=black position=center; title 'Lognormal Probability Plot for Diameters'; proc capability data=measures noprint; probplot diameter / lognormal(sigma=0.5 theta=est zeta=est color=yellow w=2) square pctlminor HREF=95 lHREF=2 hreflabel = '95%' vref = 5.8 to 6.0 by 0.1 lvref = 3 cframe = ligr legend = legend2 cHREF=red cvref = blue; run;
The plot is shown in Output 9.2.1.
Output 9.2.1: Lognormal Reference LineBased on the intersection of the diagonal reference line with the HREF=line, the estimated 95^{ th} percentile of the diameter distribution is 5.85 mm.
Note that you could also construct a similar plot in which all three parameters are estimated by substituting SIGMA=EST for SIGMA=0.5 in the preceding statements.
Chapter Contents |
Previous |
Next |
Top |
Copyright © 1999 by SAS Institute Inc., Cary, NC, USA. All rights reserved.